L3: NOTCH2NL

Question 1

What do the majority of genomic differences that make us human consist of?

Answer 1

Majority are segmental duplications

Question 2

What are segmental duplications?

Answer 2

Genomic regions greater than 1 kilobase (but usually 100s if kbs to megabases) with greater than 90% identical sequence

Question 3

How did Notch2NL come about and how conserved is it?

Answer 3

NOTCH2NL is human specific and emerged by partial segmental duplication of NOTCH2. It is located by the centromere of chromosome one. This region is highly repetitive, instable and can break easily

Question 4

Why, in the previous card, was it untrue to say NOTCH2NL gene?

Answer 4

There are multiple: NOTCH2NL genes are human specific and they emerged after a series of segmental duplications and gene conversion events involving the important neurodevelopmental gene NOTCH2. Four NOTCH2NL paralogs are present in modern humans: NOTCH2NLA, NOTCH2NLB, and NOTCH2NLC in the 1q21.1 locus and the pseudogene NOTCH2NLR next to the parental NOTCH2 gene in the 1p12 locus.

Question 5

What is the largest paralog and what does this likely mean?

Answer 5

NOTCH2NLB represents the largest duplicon in the cluster, suggesting this was the first NOTCH2NL gene present in the genome

Question 6

What is meant by copy number variation?

Answer 6

Copy number variation is a phenomenon in which sections of the genome are repeated and the number of repeats in the genome varies between individuals.

Question 7

Comment on the CNV of NOTCH2NL

Answer 7

Whereas copy-number variation is observed for NOTCH2NLC and NOTCH2NLR in the healthy human population; the copy number of NOTCH2NLA and NOTCH2NLB loci is highly stable in modern humans.

In fact, 1q21.1 copy-number variations, mediated by breakpoints within the NOTCH2NLA and NOTCH2NLB genes, are associated with various neurological disorders

Question 8

What could these findings regarding CNVs of NOTCH2NL suggest?

Answer 8

These observations suggest that the total number of functional NOTCH2NLA and NOTCH2NLB alleles may be important for normal neuronal development.

Question 9

What was Frank interested in researching as a post-doc?

Answer 9

Frank as a postdoc was interested in unique aspects of the human brain; what is the difference in the human cortex? (Similar but ours is bigger due to cortex)

Question 10

What model did Frank use to compare human and macaque brains?

Answer 10

Needed a model and developed cortical organoids for humans and macaques using embryonic stem cell cortical tissue. You add chemicals to cells to induce differentiation to replicate in-vivo as close as possible. You can mimic the layers of the cortex.

Question 11

Describe one finding he had when researching these organoids

Answer 11

Did RNA seq between human and macaque and found that results are very similar; Human and Macaque ESC cortical tissues robustly express Cortex-related genes. However, If we scroll down to specific sequences, we see that Notch2NL is not expressed in macaque, while human cortical tissues express high levels of NOTCH2NL.

Question 12

What is the relevance of NOTCH?

Answer 12

NOTCH signaling orchestrates the balance between neural stem cells and neuronal differentiation; it is conserved from fruit fly to human. Maintaining notch signalling in a cell leads to stem cell renewal. Losing notch signalling leads to premature neuronal differentiation.

Notch signaling is central to brain development, determining the timing and duration of neuronal progenitor proliferation and neuronal differentiation. It is active in outer radial glia (oRG), a cell type hypothesized to generate the majority of primate cortical neurons and to contribute to human-specific cortical expansion

Question 13

What does NOTCH2NL encode?

Answer 13

NOTCH2NL results from a partial duplication of NOTCH2. The duplicated segment includes the NOTCH2 promoter and six N-terminal epidermal growth factor (EGF)-like domains from NOTCH2 exons 1–4 but excludes the transmembrane and cytoplasmic domains. NOTCH2NL genes contain a fifth exon derived from NOTCH2 intronic sequence that provides NOTCH2NL with 20 unique amino acids. It is still secreted and in the extracellular domain.

Question 14

What did the new human genome assembly (Hg38) reveal regarding notch2nl?

Answer 14

1q21.1 was incorrectly assembled in the human reference genome until the most recent version, GRCh38. The new human genome assembly (Hg38) reveals humans have 4 nearly identical NOTCH2NL genes always in the same location

Question 15

What is significant about the 1q21 locus?

Answer 15

1q21 was subject to a large pericentric inversion involving considerable gene loss and duplication during human evolution. The 1q21 locus contains a disproportionate number of human-specific genes and also contains the 1q21.1 distal deletion and duplication syndromes interval

Question 16

What are these 1q21.1 syndromes?

Answer 16

De novo deletion of one copy frequently leads to brain size reduction (microcephaly) and duplication to brain size increase (macrocephaly), among other symptoms.

Question 17

Comment on the functionality and necessity of each of these genes

Answer 17

A and B must be present in a healthy human (fixed in the human population); D is upstream and doesn’t do much, C has low expression and is not essential.

87% of people have two C alleles, 12.5% have one and 0.2% have none.

40% of people have two D alleles, 38% have one and 21% have none.

Question 18

How similar are NOTCH2NL genes?

Answer 18

NOTCH2NL genes are mostly the same except maybe some SNPs or possibly post translational modifications. The molecular weight should be around the same (for electrophoresis).

Question 19

When analysing NOTCH2NL in human, chimp and macaque neurospheres what did they find?

Answer 19

They found Notch 2 expression in all but only NOTCH2NL in chimps.

Question 20

Why did they find higher NOTCH2NL expression in neurospheres?

Answer 20

There is higher expression in the neurospheres as they are undergoing differentiation and NOTCH2NL is important at this stage.

Question 21

What separates NOTCH2NLR from the rest making it likely to be a pseudogene?

Answer 21

In NOTCH2NLA, NOTCH2NLB, and NOTCH2NLC, the 5th exon has a 4 bp deletion compared to the corresponding sequence in NOTCH2. Mutational analysis of NOTCH2NL cDNAs shows that this 4 bp deletion is essential for NOTCH2NL protein expression

NOTCH2NLR lacks the 4 bp deletion and contains many coding variants relative to NOTCH2 and the other NOTCH2NL paralogs, including a still-segregating variant. 14% of the population lack NOTCH2NLR. Together, these results suggest but do not confirm that NOTCH2NLR is a non-functional pseudogene.

Question 22

Where do the differences between the paralogs lie?

Answer 22

NOTCH2NLA has an ATG→ATA mutation in the NOTCH2 start codon and lacks the N-terminal 39 amino acids encoding the NOTCH2 secretory pathway signal peptide.

NOTCH2NLB retains the NOTCH2 signal peptide while carrying a Thr→Ile substitution in a conserved fucosylation site.

GRCh38-NOTCH2NLC contains a 2 bp deletion just downstream of the NOTCH2 start codon and, like NOTCH2NLSh, lacks the N-terminal signal peptide.

Question 23

When did NOTCH2NL emerge?

Answer 23

The greater than 100 kb genomic regions spanning each NOTCH2NL gene show >99.1% sequence identity to NOTCH2, suggesting that NOTCH2NL paralogs were created within the last few million years, in the same time frame as SRGAP2 and HYDIN2

NOTCH2NL emerged by a partial duplication of NOTCH2 prior to the last common ancestor (LCA) of human, chimpanzee, and gorilla

Question 24

How does NOTCH2NL in gorillas and chimps compare to that of humans?

Answer 24

Unlike human, both chimpanzee and gorilla have variable read depth over the region encompassing their NOTCH2NL-like sequences. This pattern suggests the existence of multiple versions of truncated NOTCH2NL-like genes in these species. None of these genes encode NOTCH2-related proteins; they are pseudogenes.

Question 25

Describe the NOTCH2NL present in chimps

Answer 25

four chimpanzee NOTCH2NL-like pseudogenes that lack a 52 kb region including exon 2, encoding a protein of 88 amino acids with no homology to human NOTCH2NL

Four additional chimpanzee NOTCH2NL-like pseudogenes lack either exon 1 or exons 1–2 and are fused to various 3′ truncated genes

Question 26

Describe the NOTCH2NL present in gorillas

Answer 26

In gorilla, three NOTCH2NL-fusion pseudogenes were identified of which two were similar to NOTCH2NL-fusion pseudogenes found in chimpanzee: PDE4DIPexon1–27-NOTCH2NLexon2–5 and MAGI3exon1–14-NOTCH2NLexon3–5. Transcript support was obtained for BRD9-NOTCH2NL and PDE4DIPexon1–27-NOTCH2NLexon2–5 using RNA from gorilla induced pluripotent stem cells (iPSCs)

Question 27

What is likely the evolutionary history of NOTCH2NL?

Answer 27

Human NOTCH2NL genes are all in the vicinity of PDE4DIP, but human NOTCH2NL genes have a 5′ genomic structure highly similar to NOTCH2. This suggests a plausible evolutionary history of NOTCH2NL genes as follows:

Both the PDE4DIP-NOTCH2NL and MAGI3-NOTCH2NL fusion pseudogenes were present in the LCA of human, chimp, and gorilla. Then, only in the human lineage, the ancestral PDE4DIP-NOTCH2NL fusion gene was “revived” by NOTCH2 through ectopic gene conversion (recombination event that repaired the gene as it aquired the promoter that was initially present). With the acquisition of exon 1 and the upstream promoter, a viable NOTCH2NL gene encoding a stable NOTCH2-related protein was created.

Because no remnants of a MAGI3-NOTCH2NL fusion pseudogene are found in the human genome, it must have been lost, perhaps in the upheaval of the pericentric inversion on chromosome 1 and subsequent large-scale copy-number changes.

The revived human NOTCH2NL subsequently duplicated twice more to form three nearly identical NOTCH2NL genes in the 1q21.1 locus.

Question 28

Where/ when is NOTCH2NL expressed?

Answer 28

NOTCH2NL Is Expressed in Radial Glia Neural Stem Cells during Human Cortical Development. The NOTCH2NL expression pattern closely resembles that of NOTCH2 and is highest in various RG populations, including oRG as well as astrocytes and microglia

Question 29

What is the expression level of the different haplotypes?

Answer 29

NOTCH2NLB has the highest expression and that the other NOTCH2NL paratypes are expressed at levels 20%–60% of NOTCH2NLB

Question 30

What functional role does NOTCH2NL play?

Answer 30

Ectopic Expression of NOTCH2NL Delays mouse Cortical Neuron Differentiation and promotes neural stem cell proliferation; deletion in human organoids affects cortical organoid development, removing it accelerates neuron generation, promotes premature neuronal maturation and results in smaller organoids . Therefore NOTCH2NL likely helps retain stemness and increase the stem cell pool.

Question 31

How does NOTCH2NL carry out this functional role?

Answer 31

It interacts with NOTCH receptors and enhances NOTCH signal: The six N-terminal EGF-like domains encoded in NOTCH2NL, also present in NOTCH2 receptors, do not have a clearly described function in Notch signaling. Yet they are conserved from Drosophila to human, indicating an important functional role. There is evidence that N-terminal EGF-like domains are involved in dimerization of NOTCH receptors and that receptor dimerization modulates NOTCH activity (but it does not interact with EGFR and PDGFRB). Using CO-IP and luciferase assay they showed it interacts with NOTCH2 and increases its activation. This may be due to it homo-dimerising the receptor independently from ligands and exerting its effects through this manner.

Question 32

Why was NOTCH2NL never considered a candidate for the 1q21.1 disorder?

Answer 32

NOTCH2NL was never considered a candidate for the 1q21.1 distal syndrome, because in the GRCh37 assembly, which was used for the studies that mapped 1q21.1 copy-number variations (CNVs), NOTCH2NL was incorrectly positioned.

Question 33

What was the evidence for NOTCH2NL regarding the 1q21.1 syndrome?

Answer 33

They reanalysed CNV microarray data derived from 11 patients previously characterised with 1q21.1 CNV-associated microcephaly or macrocephaly by remapping to GRCh38. The results suggest that NOTCH2NLA and NOTCH2NLB are located inside the 1q21.1 distal deletion or duplication locus of these patients. All nine microcephaly cases were consistent with NOTCH2NLA and/or NOTCH2NLB deletion, and both macrocephaly cases were consistent with NOTCH2NLA and/or NOTCH2NLB duplication.

Question 34

What symptoms are associated with this 1q21.1 syndrome?

Answer 34

When present, common neurological symptoms include microcephaly (del) or macrocephaly (dup), schizophrenia (del), and autism (dup). In patients ascertained as part of the Simons Variation in Individuals (VIP) autism study, 1q21.1 distal duplication probands exhibited ADHD (29%), behavior disorder (18%), autism spectrum disorder (41%), developmental coordination disorder (23%), and intellectual disability (29%), while deletion probands exhibited these symptoms at lower frequencies, but with a relatively high percentage (26%) exhibiting anxiety and mood disorders

Question 35

What conclusions were drawn from this paper?

Answer 35

NOTCH2NL expresses a functional protein only in humans. NOTCH2NL delays cortical differentiation. The evolutionary history and function suggests a role for NOTCH2NL in the evolutionary expansion of the human neocortex

Question 36

What does the evolutionary gain of novel genes and functions can come at the expense of?

Answer 36

The evolutionary gain of novel genes and functions can come at the expense of genomic stability

Question 37

How similar are the NOTCH2NL genes?

Answer 37

NOTCH2NL genes are between 99.1-99.9% identical

Question 38

Are these differences between the NOTCH2NL genes functional?

Answer 38

Yes, they affect protein level; NOTCH2NLB creates around 5 times as much protein as A and C does not create much at all

Question 39

What difference between the genes likely correspond to this difference in protein level?

Answer 39

B has an ATG start codon, A has an ATA and multiple unconventional CTG start sites in the 5′ side of NOTCH2NL drive translation of NOTCH2NLA; C is also forced to use downstream CTG sites for translation initiation. However, due to the combination of the NOTCH2NLC-characteristic 2-bp deletion and upstream open-reading frames (ORFs), the expression level of NOTCH2NLC is extremely low, at only 1% compared with NOTCH2NLB. Therefore the level of NOTCH2NL protein generated by each of the genes is predominantly dependent on the presence or absence of three specific coding variants in Exon1

Question 40

What is meant by gene conversion?

Answer 40

Gene conversion is a specific type of homologous recombination that involves the unidirectional transfer of genetic material from a ‘donor’ sequence to a highly homologous ‘acceptor’

Question 41

What kind’ve variation is seen in the current population in regards to NOTCH2NL?

Answer 41

The number of As, Bs and Cs are variable in the population; NOTCH2NLA exon 1 configuration is represented much more than expected per individual and NOTCH2NL exon 1 configuration in the population is represented less than expected. This is striking because it suggests that the vast majority of the population carries three or four alleles with the NOTCH2NLA-derived Exon1A-(Low)-variant and only one or zero alleles with the NOTCH2NLB-derived Exon1B-(High)-variant.

Question 42

What was the exon 1 configuration like for ancient humans, denisova and neanderthals

Answer 42

In ancient humans and neanderthals the pattern for NOTCH2NLC was the same as seen in modern populations however, while denisovan measured had single unique nucleotide (SUN) reads for A, B and C (no R); It only had exons equivalent to C (low). More genomic data is required for denisova but this would appear to mean they had very low levels of NOTCH2NL protein.

Two neanderthal variants: ATG > ATA (M40I) missense variant (NOTCH2NLNea-M40I; 8%; 1 allele) and a N232S missense variant (NOTCH2NLNea-N232S; 18%; two alleles) were found in neanderthals and one denisovan variant (NOTCH2NLDen-E258A; 19%; two alleles). None of these were found in modern humans except NOTCH2NLNea-M40I very rarely (0.02%) in the human population. Nea-N232S and Den-E258A were associated with a slight decrease in the potency of NOTCH2 signalling when inserted in NOTCH2NLA. Den-E258A and Nea-N232S showed increased protein level, likely due to their effects on stability.

Question 43

What explanation was given for this finding regarding the presence of the exon 1 configurations in the population?

Answer 43

Gene conversion can surpass mendelian inheritance patterns: In a SUN analysis, they find evidence of extensive gene conversion between the NOTCH2NLA and NOTCH2NLB loci: The median SUN depth shifts in favor of either allele in different regions of the loci, indicating that parts of the NOTCH2NLA-sequence are frequently overwritten by NOTCH2NLB-sequence and vice versa (mostly intronic).

The exchange in exon 1 however shows a trend highly in favour of the NOTCH2NKA variant. This indicates that the Exon1B-(High)-variant, producing the highest levels of NOTCH2NL protein, is being lost or actively being purged out from the modern human population by gene conversion (ATG > ATA). This was also seen for ancient humans.

Question 44

When is this gene conversion likely to take place?

Answer 44

Gene conversion often takes place in Meiosis, or in early embryonic development for very unstable loci

Question 45

Despite the relatively high abundance of Exon1A-(Low) variants in NOTCH2NLA and NOTCH2NLB, some individuals still carry a relatively high number of Exon1B-(High) variants. What trend was seen in these individuals?

Answer 45

They found that individuals with a relatively high number of the Exon1B-(High) variant and low number of the Exon1A-(Low) variant often carry a nonsense SNP (R113*) in NOTCH2NLB, which leads to a premature stop-codon and a severely truncated NOTCH2NL protein

In addition, we found another variant in the splice acceptor sequence of exon 2 (Exon2B-(Splice-mut)). This variant falls outside the coding region and therefore was not detected before. The AG > GG mutation is predicted to lead to an alternative splicing event, resulting in a frameshift and truncation of NOTCH2NL proteins at amino acid 30

On hg38, the AG>GG variant is annotated in NOTCH2NLB and it is present at a high allelic frequency in human genomes. Individuals with a relatively high number of the Exon1B-(High) variant, often carry one or two alleles of the disruptive R113* mutation in NOTCH2NLB; A strikingly similar pattern was observed for the Exon2B-(Splice-mut) mutation.

Question 46

To what extent are these splice variants seen in other hominids?

Answer 46

The splice acceptor variant Exon2B-(Splice-mut) and the R113* mutation were not found in any of the currently available Neanderthal or Denisovan genomes and are therefore recently evolved human lineage-specific adaptations.

Question 47

What overall trend is seen through these mutations then?

Answer 47

Multiple variants are acting complementary to reducing NOTCH2NL protein level through mutations in NOTCH2NLB. Multiple genetic strategies reduced NOTCH2NL expression in human evolution

Question 48

What trend aligns with this trend in NOTCH2NLB variants?

Answer 48

NOTCH2NL Evolution towards reduced protein levels coincides with reduction in modern human brain size: Since ~15,000 years ago a ~10% decrease in brain size?

Question 49

What are the clinical consequences of these findings?

Answer 49

There is a high variability in symptoms are penetrance: some people with deletions have no observable features while others have variable findings. Sometimes parents carrying the full deletion have no observable features But they have affected children with the same deletion with severe NDD.

Different variants of NOTCH2NL genes might be implicated in penetrance and severity of symptoms. Parents here could could have a deletion resulting in a fusion of A and B. The starting levels of protein expression may cause the observed phenotypic variability. Fusing A to a high B may not carry much functional consequences as opposed to an A fused with a trancated B, with the splice-mut being the worst case scenario after the R113.